1 00:00:01,97 --> 00:00:05,51 Thank you all for coming out here and things to focus on 2 00:00:05,52 --> 00:00:12,26 a live show soon in. Interest of keeping things moving on time here I'll be brief 3 00:00:12,88 --> 00:00:16,13 so for the folks who haven't been in one of these lightning talk sessions yet. 4 00:00:17,19 --> 00:00:22,20 Quick overview we're going to have four minute long presentations by lightning talk 5 00:00:22,21 --> 00:00:24,96 speakers followed by four minutes of Q. 6 00:00:24,97 --> 00:00:31,12 And A for the for the speakers in the room if you can please stay within those four 7 00:00:31,13 --> 00:00:35,93 minutes just so that we can keep things on time that would be really helpful and 8 00:00:35,94 --> 00:00:42,67 for the folks in the audience who asked questions please keep your questions short 9 00:00:42,93 --> 00:00:48,84 and then for the speakers please keep your answers succinct and so we can try to 10 00:00:48,85 --> 00:00:52,30 fit as many questions as possible into that four minutes that we have for Q. 11 00:00:52,31 --> 00:00:59,10 And A with that I would like to turn it over to our host of this final lighting 12 00:00:59,11 --> 00:01:05,92 talk sessions Amy. John. Hi 13 00:01:05,93 --> 00:01:10,30 everybody my name is Amy James I'm the co-author of open index protocol which is 14 00:01:10,79 --> 00:01:12,67 open and flexible SPAC that creates 15 00:01:12,68 --> 00:01:18,70 a public space that anybody can publish to and the co-founder of Alexandria which 16 00:01:18,71 --> 00:01:21,66 is the first app built on open index protocol it's 17 00:01:21,67 --> 00:01:23,36 a search all everything that's an open and X. 18 00:01:23,37 --> 00:01:27,84 Is in there there's other projects that are building more cases and it is my 19 00:01:27,96 --> 00:01:33,80 pleasure to be introducing to the first lightning TALK OF THE DAY J grabber who 20 00:01:33,81 --> 00:01:37,99 will be demo ing Twitter of patchwork sorry 21 00:01:38,09 --> 00:01:45,02 a demo of Twitter to Patchwerk bridge so here she here. Sorry sorry sorry 22 00:01:45,03 --> 00:01:50,73 I didn't know you are going here. Right so I'm going to talk on social network 23 00:01:50,74 --> 00:01:54,60 interoperability as something I think we need for decentralized social networks. 24 00:01:56,00 --> 00:01:59,73 And I was going to originally talked just about the patchwork to Twitter 25 00:01:59,74 --> 00:02:03,03 integration I made but since then I kind of started hacking on this more as 26 00:02:03,04 --> 00:02:07,35 a side project so I'm going to talk about the concept in general and what I think 27 00:02:07,62 --> 00:02:11,59 it's possible with it and show off the Twitter Patrick bridge and since we judge 28 00:02:11,60 --> 00:02:14,36 people OK So right now 29 00:02:14,37 --> 00:02:18,14 a lot of these like decentralized social networks exist basically as little islands 30 00:02:18,15 --> 00:02:22,41 that are totally disconnected from the civilizations of Twitter and Facebook and 31 00:02:22,50 --> 00:02:25,76 although I want to use them I end up on the platform because they're easy to use 32 00:02:25,77 --> 00:02:30,96 and all people I talk to are there and so in order to get more adoption in addition 33 00:02:30,97 --> 00:02:34,53 to better you x I think we need more interoperability between centralized platforms 34 00:02:34,54 --> 00:02:37,54 and these decentralized networks so there's 35 00:02:37,55 --> 00:02:40,58 a thing called the Data Transfer project which Google just open source some code 36 00:02:40,59 --> 00:02:41,83 for recently is 37 00:02:42,03 --> 00:02:47,37 a kind of interoperable protocol for these big platforms and designs that you 38 00:02:47,38 --> 00:02:50,100 transfer data between them but it's built more for like platforms and sell them for 39 00:02:51,01 --> 00:02:55,66 service providers rather than for individuals but what I kind of want is that but 40 00:02:55,67 --> 00:02:58,86 for the decentralized Web so you have something in between or like 41 00:02:58,87 --> 00:03:03,78 a library that will convert for you between you know your data from these networks 42 00:03:03,84 --> 00:03:09,26 into whichever format you want to use. So sometimes I think that 43 00:03:09,27 --> 00:03:12,79 a centralized web applications could take towards interoperability you could set up 44 00:03:12,97 --> 00:03:17,03 ways to help users import their data from existing platforms help them authenticate 45 00:03:17,04 --> 00:03:21,46 their identity and let them link back to you know their Twitter profile their 46 00:03:21,47 --> 00:03:25,13 Facebook profile things like that and allow you to post back those platforms if you 47 00:03:25,14 --> 00:03:26,05 have a user's A.P.I. 48 00:03:26,06 --> 00:03:31,26 He's been authorized the application to do that. And then you can also lead you 49 00:03:31,27 --> 00:03:34,16 through the streams what they want to hold on to and what to discard like maybe 50 00:03:34,17 --> 00:03:34,77 make them like 51 00:03:34,78 --> 00:03:39,83 a place to aggregate their information and discard what they don't want and kind of 52 00:03:39,84 --> 00:03:41,08 like to curate it in 53 00:03:41,29 --> 00:03:45,34 a better way. So I think one thing to be necessary to do this would be 54 00:03:45,35 --> 00:03:48,61 a library for data conversions and the sort of your social format kind of like the 55 00:03:48,62 --> 00:03:53,75 data transfer project for the web. So when I first started doing this it was 56 00:03:53,76 --> 00:03:57,27 because I used patchwork with essentially social networks and I use Twitter and I 57 00:03:57,28 --> 00:04:01,15 wanted I just wanted some communication between the two so I hacked my patchwork 58 00:04:01,16 --> 00:04:03,59 client to post a Twitter you can see over here the post 59 00:04:03,60 --> 00:04:08,85 a Twitter button when you post that message it goes to this test test Twitter 60 00:04:08,86 --> 00:04:12,91 account I made which posted password and I made 61 00:04:12,92 --> 00:04:15,40 a test run because of course I like posted like test 62 00:04:15,41 --> 00:04:22,09 a bunch of times and it was kind of spamming my feed but so yeah it is possible and 63 00:04:22,13 --> 00:04:26,01 that's how I work and then you can do a Twitter patchwork import so as 64 00:04:26,02 --> 00:04:29,63 a user I can request my data from Twitter I got the face on like 65 00:04:29,64 --> 00:04:32,59 a dump of all my tweets and then I wrote 66 00:04:32,60 --> 00:04:36,88 a script to import it into your skull but my password client so now this is an 67 00:04:36,89 --> 00:04:37,42 example of 68 00:04:37,43 --> 00:04:43,65 a patchwork post that links to my Twitter original tweet on that from Twitter and 69 00:04:43,66 --> 00:04:48,76 so here's the original tweet and so with that than I So we have some integration 70 00:04:48,77 --> 00:04:52,37 between these different identities and you can like take this 71 00:04:52,38 --> 00:04:56,64 a lot further but for patchwork or for what other social media client I think 72 00:04:56,65 --> 00:05:01,17 you're building so then I was kind of like just working with those trying to make 73 00:05:01,18 --> 00:05:06,83 a little parser for these things to start calling it bridge. And eventually would 74 00:05:06,84 --> 00:05:10,43 be great to have library for this right now it's just some day some parsing for 75 00:05:10,44 --> 00:05:15,24 Facebook and Twitter formats. But then I started working on 76 00:05:15,25 --> 00:05:19,40 a mobile app says like so much of the decentralized web is built for permanence and 77 00:05:19,41 --> 00:05:24,76 built for computers and many things really usable So here's an example of reactive 78 00:05:24,77 --> 00:05:26,72 mobile apps I started building which does 79 00:05:26,73 --> 00:05:32,59 a same thing gives you posts Twitter Facebook native posts links back to your 80 00:05:32,60 --> 00:05:34,63 networks. Gives you 81 00:05:34,64 --> 00:05:39,81 a full view that lets you delete the post and view the original and editing 82 00:05:39,82 --> 00:05:43,91 intuition something else I think that social media should be ephemeral and so 83 00:05:43,95 --> 00:05:47,47 essential a social network that focus on permanence are I think going 84 00:05:47,48 --> 00:05:53,55 a bit in the wrong direction so all the code. Here Patrick where Twitter S.S.P. 85 00:05:53,65 --> 00:05:58,32 Bridge is the place on pricing bridge mobile was the app and bridge serve as the Go 86 00:05:58,33 --> 00:06:02,88 servers for that and I mark when you're on Twitter past work and write if you want 87 00:06:02,89 --> 00:06:03,92 to talk more about it. 88 00:06:11,96 --> 00:06:12,73 Any question. 89 00:06:21,46 --> 00:06:24,92 Do you envision that people would run this themselves or do you think that there 90 00:06:24,93 --> 00:06:26,18 would be somebody running this for 91 00:06:26,19 --> 00:06:30,29 a whole bunch of people how do you think this would actually be operationalized I 92 00:06:30,30 --> 00:06:32,68 invented being a library where anyone creating 93 00:06:32,69 --> 00:06:38,70 a decentralized social network could use to start making those conversion. Or like 94 00:06:38,74 --> 00:06:45,74 a. Like an application make an aggregator that does this. Yeah 95 00:06:45,80 --> 00:06:48,37 just like different versions of this but the general principle should be as 96 00:06:48,38 --> 00:06:49,70 a reality and interoperability. 97 00:07:02,61 --> 00:07:06,43 But does this fit in with the activity pub are you publishing to 98 00:07:06,45 --> 00:07:12,59 a certain data model. No I didn't so I don't really have like the in-between data 99 00:07:12,59 --> 00:07:17,99 model figured out exactly but I think that you should publish to a year to have 100 00:07:18,01 --> 00:07:22,59 a library that like converts into all these different formats right so you have 101 00:07:22,60 --> 00:07:29,53 actively PA like your scuttlebutt matrix. I'm sure there's others they want 102 00:07:29,54 --> 00:07:33,30 to go to like do those conversion for as many as you can add in like plug and 103 00:07:33,34 --> 00:07:34,89 basically like like 104 00:07:34,90 --> 00:07:39,60 a new Web there's that Bridgie which does something similar Yes of course. 105 00:07:51,19 --> 00:07:57,77 I've used bridging stuff with. Diaspora and 106 00:07:57,78 --> 00:08:02,74 a couple of other decently social networks at Bill something like this where I 107 00:08:02,75 --> 00:08:08,70 always noticed is that the formatting can be kind of wonky on either side like the 108 00:08:08,71 --> 00:08:12,99 message might be mingled in some way like maybe it's cut off half way and there's 109 00:08:13,00 --> 00:08:13,10 like 110 00:08:13,11 --> 00:08:16,25 a link that you have to click to really read the rest of the thing or something like 111 00:08:16,26 --> 00:08:23,12 that is do you have it right now is. Set up so that it's you know it 112 00:08:23,13 --> 00:08:23,48 looks like 113 00:08:23,49 --> 00:08:30,32 a native post as opposed to something that's clearly not. Easiest because it's 114 00:08:30,33 --> 00:08:31,54 just for 115 00:08:31,55 --> 00:08:36,06 a short text but yeah of course I think some you have to do some integration or 116 00:08:36,07 --> 00:08:41,35 whatever it's like. I actually I don't you know I think I've learned more about 117 00:08:41,36 --> 00:08:45,32 activity pub at this conference I don't know everything about his or format I work 118 00:08:45,33 --> 00:08:48,20 mostly as an engineer. By project. 119 00:08:50,52 --> 00:08:55,39 Thank you thank. 120 00:08:58,68 --> 00:09:03,01 You I've been he's going to talk about his project we met a cooperative P.D.P. 121 00:09:03,27 --> 00:09:06,76 Cash that helps overcome network interface figure. 122 00:09:11,55 --> 00:09:17,12 Thank you. So. I'm going to do that which is basically 123 00:09:17,80 --> 00:09:24,02 a technology that allows users to. Overcome 124 00:09:24,67 --> 00:09:31,61 connectivity issues with upstream web servers so what's the 125 00:09:31,62 --> 00:09:38,49 reason I brought in that we want to do. So. Tim Berners 126 00:09:38,50 --> 00:09:44,50 Lee already. First ever web server OK as you will see here it has 127 00:09:44,51 --> 00:09:51,46 a sticker we. Do not turn out. OK So Tim Berners Lee already knew 128 00:09:51,47 --> 00:09:54,89 that the certain web server not being available the problem. 129 00:09:59,28 --> 00:10:05,59 In the for about two weeks for the fun let's be fair. OK 130 00:10:06,18 --> 00:10:13,01 They're sold Fortunately decentralisation Well the previous problem is 131 00:10:13,02 --> 00:10:18,58 because the servers Web servers spread play as centralized role in the Web So this 132 00:10:18,59 --> 00:10:24,89 is really Satan personally have had several issues several features that helps 133 00:10:25,22 --> 00:10:31,13 overcome that and I was not going to explain them here or really know them OK so 134 00:10:31,14 --> 00:10:31,36 with 135 00:10:31,71 --> 00:10:39,49 a different I say front we're seeing really fantastic new Nobody solutions. However 136 00:10:39,50 --> 00:10:42,12 the some some of them just walk 137 00:10:42,20 --> 00:10:48,87 a partial or walled away of the established technologies so much that users 138 00:10:49,19 --> 00:10:52,86 frequently have issues normal normal users can have 139 00:10:52,87 --> 00:10:57,97 a hard time using them out they at least need to learn many new concepts and ways 140 00:10:57,98 --> 00:11:04,37 of using the. New technologies so what we want with you when it 141 00:11:04,77 --> 00:11:09,90 is to provide something that if completely seamless unfamiliar to the normal web 142 00:11:09,91 --> 00:11:15,27 user to can just use the same tools the browser the concepts as you are else for 143 00:11:15,28 --> 00:11:22,02 content providers they can use the normal web servers we want to to have that. Tool 144 00:11:22,30 --> 00:11:28,86 don't exclude with your services or. Or require for from them other 145 00:11:28,90 --> 00:11:35,88 special knowledge or or even if you. Feel that we want complicity in 146 00:11:35,89 --> 00:11:42,78 familiar browsing OK so what we need. When it 147 00:11:42,79 --> 00:11:44,62 is basically called currently 148 00:11:44,63 --> 00:11:51,54 a library. That extends where were public patience accessing the web 149 00:11:52,11 --> 00:11:59,11 with new worlds and try superpowers peer to peer routing and distributed caching 150 00:11:59,62 --> 00:12:05,31 which is provided by users themselves OK. From the point of view of content 151 00:12:05,49 --> 00:12:11,78 providers basically Labor says became if you could see the end of the of the blue. 152 00:12:14,42 --> 00:12:21,32 So typical The use case and when it is that somebody is 153 00:12:21,33 --> 00:12:27,43 visiting. Some content like this on the most piece of news. 154 00:12:29,35 --> 00:12:36,34 On their normal browsers. When it slavery at the same time up lol The 155 00:12:36,42 --> 00:12:42,21 content to the distributive cache. So that if something happens. 156 00:12:43,32 --> 00:12:50,01 And we start having problems with written the the afternoon web server because this 157 00:12:50,01 --> 00:12:55,29 causes the for normal web users to not be able to receive content then with Wynette 158 00:12:55,29 --> 00:12:57,75 . If somebody's 159 00:12:57,76 --> 00:13:02,18 a ready access this this web page they can retrieve the content from from other 160 00:13:02,19 --> 00:13:09,04 users even if it may be outdated because maybe the. Latest freshest 161 00:13:09,04 --> 00:13:13,31 content is not available because there has been this this issue with absolute 162 00:13:13,32 --> 00:13:19,86 connectivity they at least get something. OK So in some 163 00:13:19,87 --> 00:13:24,46 countries international connectivity is quite slow in their use so may suffer from 164 00:13:24,47 --> 00:13:30,92 from that but not only that from time to time there's no neutral pricing of 165 00:13:30,96 --> 00:13:36,01 international traffic which makes it overly expensive and this can be 166 00:13:36,39 --> 00:13:39,94 a better price for who says we're paying the the A.P. 167 00:13:39,95 --> 00:13:44,66 Bill. So we've had. Since 168 00:13:44,67 --> 00:13:51,43 a. WE ARE YOU were retrieving content from differend different uses that made the 169 00:13:51,47 --> 00:13:58,47 faster. So that we ship typical. So how 170 00:13:58,48 --> 00:14:05,37 does it work. When the request from the from there because some powered by 171 00:14:05,38 --> 00:14:09,13 when it goes to the to the local client it decides whether to send it to the origin 172 00:14:09,14 --> 00:14:11,62 server or to a proxy of 173 00:14:11,63 --> 00:14:16,15 a peer to peer writing or to the disputed gas if it's not there it contacts 174 00:14:16,15 --> 00:14:16,95 a trusted with 175 00:14:17,19 --> 00:14:22,39 a trusted injector which uploaded to the to the network it's completely 176 00:14:22,40 --> 00:14:27,58 configurable it's more complex than that the sun the in the works and OK 177 00:14:27,88 --> 00:14:32,45 technically zero for it OK fine they are. 178 00:14:34,88 --> 00:14:41,79 OK. For the future of the film is to do the work on supporting 179 00:14:41,80 --> 00:14:47,93 more backhand and that there was billions to do with censorship in theatre not 180 00:14:47,94 --> 00:14:52,47 likin private browsing also and keep an eye on resource usage for use in on mobile 181 00:14:52,48 --> 00:14:55,60 phones. But that you have questions 182 00:14:58,93 --> 00:15:03,76 thanks. 183 00:15:09,31 --> 00:15:15,77 For. More into the chain about privacy 184 00:15:16,34 --> 00:15:20,20 and how if it would you say there were issues around privacy thing we know because 185 00:15:20,24 --> 00:15:25,37 you are you are applauding what you with your browsing history fuel or your 186 00:15:25,38 --> 00:15:27,94 browsing data what the pages I did download it to 187 00:15:27,95 --> 00:15:34,65 a disputed public cache so one must be. Careful when deciding OK this content I can 188 00:15:34,66 --> 00:15:39,95 consider it public and it's fine if published or this is not I don't intend this 189 00:15:40,15 --> 00:15:44,63 one this is the to be published so it's completely configurable by the user about 190 00:15:45,27 --> 00:15:49,91 why and need to be careful that's why we consider to feel Alpha and not as ready 191 00:15:49,92 --> 00:15:51,80 for everyone everybody to use. 192 00:16:02,64 --> 00:16:09,19 OK thank you very much thank. OK There we go. 193 00:16:10,98 --> 00:16:16,48 So I'm going to give an overview of the sign at work today. And so size is 194 00:16:16,49 --> 00:16:19,06 a decentralized cloud storage platform it's 195 00:16:19,07 --> 00:16:23,42 a place where you can upload your data. It aims to be similar to 196 00:16:23,76 --> 00:16:29,71 a detention as version of Amazon as three. But I want to put it in context of some 197 00:16:29,72 --> 00:16:34,53 of the other like decentralized data protocols that exist I think two good examples 198 00:16:34,54 --> 00:16:41,16 would be Bit Torrent and IP FS which are decentralized and I guess sort of file 199 00:16:41,17 --> 00:16:47,03 sharing platforms that maybe that may be too specific but they both suffer from 200 00:16:47,04 --> 00:16:52,65 a big issue that they cannot take files which are not meant to be social in nature 201 00:16:53,00 --> 00:16:56,03 so on both both Bit Torrent and IP I pass 202 00:16:56,37 --> 00:17:00,53 a file that can only be retrieved if someone is sitting there seeing it and that is 203 00:17:00,54 --> 00:17:05,19 generally an altruistic behavior there's no reason for someone to take data and 204 00:17:05,20 --> 00:17:10,34 seed it and so this works very well for public files or files that have like social 205 00:17:10,35 --> 00:17:15,36 value to them but it doesn't work for private files for example if I want to back 206 00:17:15,37 --> 00:17:19,96 up my computer I'm probably going to encrypt it before I posted in 207 00:17:20,01 --> 00:17:23,64 a public spot and then if I put it on bittorrent right to pass no one's going to 208 00:17:23,65 --> 00:17:29,08 see it so if my machine goes down that backups not worth anything to me so either I 209 00:17:29,09 --> 00:17:33,88 have to make physical backups and worry about distributing it manually or I have to 210 00:17:33,89 --> 00:17:34,08 use 211 00:17:34,09 --> 00:17:41,09 a centralized service so CYA sort of solves this problem there is no nobody to 212 00:17:41,35 --> 00:17:47,25 host your data and it solves this problem by using incentives where bittorrent and 213 00:17:47,26 --> 00:17:53,34 IP Fs are free networks. Because of the social aspect you can expect people to see 214 00:17:53,38 --> 00:17:58,19 certain types of files for free you don't have to worry about it. Side of the page 215 00:17:58,20 --> 00:18:02,80 network so if you put data on the sign at work you're going to be paying not one 216 00:18:02,81 --> 00:18:03,27 person but 217 00:18:03,28 --> 00:18:06,94 a bunch of different people to hold that trial and because they're being paid they 218 00:18:06,95 --> 00:18:08,54 don't care what it is they don't need 219 00:18:08,55 --> 00:18:13,91 a reason to see that the money is is reason enough to see that. And then beyond 220 00:18:13,92 --> 00:18:15,97 that. We actually have 221 00:18:15,98 --> 00:18:22,52 a collateral system so we don't trust our hosts we don't want to fall back to 222 00:18:22,53 --> 00:18:26,98 a centralized model like Amazon where where we have these like trust contracts in 223 00:18:26,99 --> 00:18:32,37 place and that we just have to depend on Amazon and we want to be free of that so 224 00:18:32,38 --> 00:18:33,08 instead of using 225 00:18:33,09 --> 00:18:39,09 a name like Amazon dot com We use an incentive system so before we give files to 226 00:18:39,10 --> 00:18:41,53 someone else to hold onto them we create 227 00:18:41,54 --> 00:18:45,52 a contract that we put money into That's reward for them if they hold on to the 228 00:18:45,53 --> 00:18:49,80 files but we also require that they put money into the contract so if they want 229 00:18:49,81 --> 00:18:54,57 your revenue they have to take money out of their own pocket put in to put it into 230 00:18:54,58 --> 00:18:58,17 the contract as a sign of good faith and that contract goes on 231 00:18:58,18 --> 00:19:04,30 a block and so there's no centralized escrow enforcing this contract is handled 232 00:19:04,31 --> 00:19:07,26 it's handled by block and then this contract has 233 00:19:07,98 --> 00:19:13,29 a time like associated with it so the money's going to be locked up for say twelve 234 00:19:13,30 --> 00:19:17,76 weeks at the end of the twelve weeks the person holding your data has to prove that 235 00:19:17,77 --> 00:19:22,21 they still have the data that they fulfilled the contract if they do that they get 236 00:19:22,22 --> 00:19:26,83 their money back they get the money that you promised them and they don't need you 237 00:19:26,84 --> 00:19:30,89 to be there in order to claim it so you can go on vacation go missing or whatever 238 00:19:31,13 --> 00:19:34,04 the person holding your data is going to continue holding their data because the 239 00:19:34,05 --> 00:19:37,61 payment is guaranteed and Similarly you don't have to worry about it because you 240 00:19:37,62 --> 00:19:42,86 know that if they disappear they're going to drop your pile. And like you said we 241 00:19:42,87 --> 00:19:48,09 don't trust just one person things happen fires happen whatever so what we do is 242 00:19:48,10 --> 00:19:51,89 for you something called green Solomon coding we can take this file we can split it 243 00:19:51,90 --> 00:19:57,36 up into thirty pieces we can make it so out of thirty pieces we only need ten and 244 00:19:57,37 --> 00:20:02,30 doesn't matter which ten as long as we get some ten pieces we can recover the 245 00:20:02,31 --> 00:20:06,26 original file and erase your codes are kind of magical in that this only causes 246 00:20:06,27 --> 00:20:07,10 three X. 247 00:20:07,43 --> 00:20:13,51 Overhead. That's all I can explain in four minutes but I hope you find it interesting 248 00:20:13,51 --> 00:20:19,91 . Thank you I was. 249 00:20:21,27 --> 00:20:25,73 Questions which are 250 00:20:31,73 --> 00:20:38,00 things for company given that. Was the application ecosystem look like for SEO in 251 00:20:38,01 --> 00:20:41,86 terms of user interfaces Yeah so we have 252 00:20:41,87 --> 00:20:47,53 a couple applications the most prominent is probably called my inbox which is 253 00:20:47,54 --> 00:20:53,41 a and it was acquired by Clear Center it's a nasty solution. So 254 00:20:53,42 --> 00:21:00,26 a box that you can put storage and so it has extra space on it most Nasa's 255 00:21:00,27 --> 00:21:05,77 are typically mostly on use that extra space gets sold over the sign at work so you 256 00:21:05,78 --> 00:21:09,83 can make money on that extra space and then the space that is use gets backed up to 257 00:21:09,84 --> 00:21:13,30 the sign that work so that if something happens locally you have 258 00:21:13,32 --> 00:21:19,39 a way to recovery data and we also have a called Q box which is trying to be 259 00:21:19,39 --> 00:21:23,44 a decentralized version of Dropbox and so tries to mimic the you X. 260 00:21:23,46 --> 00:21:28,53 And feeling easiness of Dropbox but instead of giving it to Dropbox the servers 261 00:21:28,75 --> 00:21:30,72 it's on this nice decentralized platform. 262 00:21:36,66 --> 00:21:42,65 It's. Zero on and that it can load down load balancing or like you know like when 263 00:21:42,66 --> 00:21:46,62 the road goes out so but you still want to see them out of redundancy so another 264 00:21:46,63 --> 00:21:53,43 one takes up the contracts and. Yes there is auto repair this is not network side 265 00:21:53,64 --> 00:21:59,64 the repair and the reason is the duplication so if the network knows how to 266 00:22:00,61 --> 00:22:05,89 shrink and grow the redundancy of your file it's possible for hosts to reverse that 267 00:22:05,90 --> 00:22:11,58 figure out how to cheat. Your own client has to manage the redundancy and re 268 00:22:11,59 --> 00:22:16,21 replication but it does do that automatically and so we generally say as long as 269 00:22:16,22 --> 00:22:22,72 your client is running. A couple nights overnight per month it should be able 270 00:22:22,73 --> 00:22:29,26 to. Manage Any house that offline it can repair your files that have weakened 271 00:22:29,27 --> 00:22:32,55 redundancy and so you will you know restore them and of course the host the do go 272 00:22:32,56 --> 00:22:34,62 offline get analyzed for it. 273 00:22:42,52 --> 00:22:49,37 Course. Uses something like that back in the web say so instead of you know using S 274 00:22:49,38 --> 00:22:54,36 three a developer would use with their put user content on C.N.N. 275 00:22:54,37 --> 00:22:58,41 At work and you can load from the with browser Yes So everything I've described up 276 00:22:58,42 --> 00:23:02,61 to this point is stuff that's working today you can go download and use it Website 277 00:23:03,04 --> 00:23:07,16 support is something we have planned in the future. Or C.D.N. 278 00:23:07,17 --> 00:23:10,33 Support really so we would we would envision something like a site like 279 00:23:10,34 --> 00:23:17,26 a major instead of streaming images or if it's Netflix you know movies from their 280 00:23:17,49 --> 00:23:22,40 central infrastructure the movies just set on the sign network the user has no idea 281 00:23:22,41 --> 00:23:22,80 that there's 282 00:23:22,81 --> 00:23:26,32 a decentralized network involved they just go to Netflix and start streaming 283 00:23:26,33 --> 00:23:30,88 a movie and it hits them from hosts on the sign network and of course this is this 284 00:23:30,89 --> 00:23:30,94 is 285 00:23:30,95 --> 00:23:35,45 a much cheaper much faster and more efficient than the solutions that Netflix uses 286 00:23:35,46 --> 00:23:39,43 today but that's not deployed that is something that is in development. 287 00:23:45,89 --> 00:23:52,81 What are the costs like for users and what are the rewards 288 00:23:52,82 --> 00:23:59,82 like sure so. Amazon typically charges twenty one 289 00:23:59,83 --> 00:24:04,39 dollars per terabyte per month depending on the volume on the sign network for 290 00:24:04,40 --> 00:24:09,20 storage it costs about two dollars per terabyte per month that's afford or 291 00:24:09,21 --> 00:24:11,82 magnitude cheaper at three X. Or done in C. 292 00:24:12,29 --> 00:24:16,35 That means that hosts are getting about paid about seventy cents per terabyte per 293 00:24:16,36 --> 00:24:21,24 month so they see one third of that each of the redundancy band with Amazon is 294 00:24:21,25 --> 00:24:25,07 typically around ninety dollars per terabyte although in bulk it goes down to fifty 295 00:24:25,08 --> 00:24:28,38 dollars per terabyte now Messiah network I believe right now it's about fifteen 296 00:24:28,64 --> 00:24:35,26 cents per terabyte so it's practically zero and that's all the time we have thanks 297 00:24:35,27 --> 00:24:37,80 very much. Thank 298 00:24:42,58 --> 00:24:48,66 you so much that was super interesting I have heard that. Free hard drive space on 299 00:24:48,67 --> 00:24:53,76 all of our computers is actually more than if you combined all the CD ends of all 300 00:24:53,77 --> 00:24:58,76 of those major corporations you're talking about is that right yeah yeah that's 301 00:24:58,77 --> 00:25:05,30 pretty cool and. You get to go 302 00:25:05,59 --> 00:25:09,76 Yeah right this is still fun Thomas and he is going to be talking about web 303 00:25:09,77 --> 00:25:16,08 monetization. Right. Yes can you me 304 00:25:16,99 --> 00:25:21,96 so yeah I'm seven Thomas I've been watching community for about eight years and 305 00:25:21,97 --> 00:25:25,69 today I'm going to talk about the current project called my salvation which is sort 306 00:25:25,70 --> 00:25:25,77 of 307 00:25:25,78 --> 00:25:30,88 a new proposed web standard. But first we're going to explain the problem like if you're 308 00:25:30,89 --> 00:25:37,20 browsing the web today you probably run into this kind of screen which is. You know 309 00:25:37,21 --> 00:25:42,52 most websites nowadays if newspapers or even like any kind of free service there's 310 00:25:42,53 --> 00:25:42,77 going to be 311 00:25:42,78 --> 00:25:46,94 a lot of advertising mixed in with your content in fact if you kind of color it you 312 00:25:46,95 --> 00:25:50,31 can see like all the red areas are advertising with like the whole screen is taken 313 00:25:50,32 --> 00:25:53,35 up with just ads and so it's not surprising that 314 00:25:53,36 --> 00:25:57,43 a lot of users are actually using ab locker sound and employers are growing almost 315 00:25:57,44 --> 00:26:00,74 exponentially. But that's also not really 316 00:26:00,75 --> 00:26:04,83 a solution because the as the web that's actually relying on those ads to make 317 00:26:04,84 --> 00:26:09,92 revenue you can't really pay journalists you know unless you have some revenue so 318 00:26:09,93 --> 00:26:14,09 how do we solve that problem and you know from the from the perspective of the Web 319 00:26:14,10 --> 00:26:17,83 site like You shouldn't be able to consume the content unless you're in some way 320 00:26:17,84 --> 00:26:24,78 contributing to it but really does it have to be advertising. From the user 321 00:26:24,79 --> 00:26:29,49 side like I want to contribute pretentiously right like most people that use ad 322 00:26:29,61 --> 00:26:33,60 blockers when they're asked they actually say that I feel guilty about using an ad 323 00:26:33,61 --> 00:26:37,10 blocker and I think the rest they would still say that they want to support the 324 00:26:37,11 --> 00:26:41,06 website they just isn't a good way and so they feel like they don't have 325 00:26:41,07 --> 00:26:47,10 a good option and so the way we do it today is kind of like this where if I use I 326 00:26:47,11 --> 00:26:49,97 go to websites I get basically the content back and I get 327 00:26:49,98 --> 00:26:53,34 a bunch of stuff that I don't want which is the ads and that's how the Web site 328 00:26:53,35 --> 00:26:57,47 makes money. And you know with any other kind of business whether it's 329 00:26:57,48 --> 00:26:58,41 a grocery store or 330 00:26:58,42 --> 00:27:02,42 a restaurant or anything else what you would do is more something like this where 331 00:27:02,63 --> 00:27:07,14 you know you give me something useful I give you some money in return that's it you 332 00:27:07,15 --> 00:27:12,40 know and what's funny is that actually the creators of the web actually envisioned 333 00:27:12,41 --> 00:27:15,68 that it would work that way so there's a status code reserved in the H.T.T.P. 334 00:27:15,69 --> 00:27:20,39 Standard for paying for the Web site or like you have to pay for this particular 335 00:27:20,40 --> 00:27:22,22 page and they just never had 336 00:27:22,23 --> 00:27:25,69 a way to actually realize this and Stuart twenty years later and or thirty years 337 00:27:25,70 --> 00:27:30,80 later and we still don't have this. And so we kind of I was involved in an open 338 00:27:30,81 --> 00:27:34,73 source project called into leisure It's a community group at the W three C. 339 00:27:34,74 --> 00:27:39,04 It's actually the third largest community group there behind Web assembly and 340 00:27:39,16 --> 00:27:43,35 progressive Web apps and it's got contributors from all kinds of backgrounds from 341 00:27:43,36 --> 00:27:47,22 the boxing community but also from so much additional finance communities in tech 342 00:27:47,23 --> 00:27:52,18 community even like some regulators and stuff like that are in there and basically 343 00:27:52,19 --> 00:27:56,31 with this committee has been has come up with this this protocol which essentially 344 00:27:56,32 --> 00:28:01,49 abstracts away how money moves so rather than saying like oh we're going to use 345 00:28:01,65 --> 00:28:02,66 payment method X. 346 00:28:02,67 --> 00:28:06,77 Whether that axis Bit Coin or Pay Pal we're just going to say like how we're going 347 00:28:06,78 --> 00:28:08,78 to use into ledger and that could go through X. 348 00:28:08,79 --> 00:28:12,86 And pay Paul and something else and yet another network just like on the Internet 349 00:28:12,87 --> 00:28:13,63 it might go 350 00:28:13,84 --> 00:28:19,75 a packet might go over wife I and then Ethan that and and so on. And so now I want 351 00:28:19,76 --> 00:28:23,31 to show you a quick demo of what that looks like so here I'm in 352 00:28:23,32 --> 00:28:25,94 a Incognito tab and I'm looking at a Web site the house of 353 00:28:25,95 --> 00:28:30,07 a photo gallery and in this web so you don't see any ads and there's some free 354 00:28:30,08 --> 00:28:33,41 content so I can look at a picture and obviously insisted it's going to take 355 00:28:33,42 --> 00:28:37,51 a second to load but I can look at different pictures and that's fine but then when 356 00:28:37,52 --> 00:28:37,99 I want to look at 357 00:28:38,00 --> 00:28:43,13 a picture that is smart this paid content I actually won't be able to look at it 358 00:28:43,32 --> 00:28:49,16 and so it says OK I need to wear my salvation enabled browser and so over here I 359 00:28:49,17 --> 00:28:49,37 have 360 00:28:49,38 --> 00:28:54,88 a browser that is logged into what my station account and is able to pay and so now 361 00:28:54,89 --> 00:29:01,73 I can like look at the Paid take your and it should hopefully works. If you pay 362 00:29:01,74 --> 00:29:06,65 for it and then show me the picture and so what's happening here is that it's 363 00:29:06,66 --> 00:29:10,99 actually the Web site telling the browser please pay to this end point and then the 364 00:29:11,00 --> 00:29:11,51 browser makes 365 00:29:11,52 --> 00:29:15,11 a paid request that is and point and then retrieve the picture from there so that 366 00:29:15,12 --> 00:29:18,03 endpoint doesn't even have to belong to that Web site so you can even have like 367 00:29:18,07 --> 00:29:23,46 apps that you know pay other payday P I's and things like that so yeah so I work 368 00:29:23,47 --> 00:29:27,11 for a company that is essentially we try to make 369 00:29:27,12 --> 00:29:31,77 a flat rate to give you this kind of functionality but the protocols are all open 370 00:29:31,78 --> 00:29:35,57 so if you want to make your own provider for this or you want to make an app that 371 00:29:35,63 --> 00:29:39,10 charges money through weaponization you can. You can. 372 00:29:47,09 --> 00:29:53,58 Any question about. So when use 373 00:29:53,59 --> 00:29:55,67 appears that something is then 374 00:29:55,68 --> 00:30:01,48 a guarantee from somewhere that the content would be served it was. Supposed to be 375 00:30:01,52 --> 00:30:05,59 behind people and yeah great question so I think that with 376 00:30:05,60 --> 00:30:10,18 a lot of these kinds of interactions the amounts are so small that the friction you 377 00:30:10,19 --> 00:30:17,01 introduce from adding too many like checks and extra things actually 378 00:30:17,02 --> 00:30:22,39 outweigh the value of the transaction itself of kind of like we use the amount of 379 00:30:22,40 --> 00:30:23,64 time you spend on the Web site as 380 00:30:23,65 --> 00:30:27,81 a proxy for how much value get out of it and I think the future might come up with 381 00:30:27,82 --> 00:30:33,56 some better metrics but basically if you click on the Web site and it's like 382 00:30:33,57 --> 00:30:36,99 a click baby Web site and you don't get much value out of you to click away as it 383 00:30:37,00 --> 00:30:40,55 was I would make very much money but if it's an app you keep coming to it over time 384 00:30:40,56 --> 00:30:40,88 you would make 385 00:30:40,89 --> 00:30:45,93 a lot more and we think that that forces enough discipline to kind of provide 386 00:30:45,94 --> 00:30:46,48 a good service. 387 00:30:53,47 --> 00:30:59,70 Yeah so yes answer so the question is how does that compare to braise approach so 388 00:30:59,74 --> 00:31:03,72 right now bravest It's like a platform so if you're 389 00:31:03,73 --> 00:31:06,49 a publisher you have to actually register you have to go break the comics or 390 00:31:06,50 --> 00:31:07,87 register you have to have 391 00:31:07,88 --> 00:31:10,68 a relationship with Bryce to get paid whereas here there's 392 00:31:10,69 --> 00:31:14,59 a protocol in the middle so you can essentially just call an A.P.I. 393 00:31:14,60 --> 00:31:19,69 In the browser and the side that is calling it could be any publisher just calling 394 00:31:19,70 --> 00:31:24,04 it just go function and the side that's paying could be any provider so you could 395 00:31:24,08 --> 00:31:27,30 it's not tied to one particular platform. 396 00:31:32,16 --> 00:31:33,82 Yeah so on the protocol level it's just 397 00:31:33,83 --> 00:31:38,42 a paywall I think that on the user side you may want to have 398 00:31:38,43 --> 00:31:41,61 a different experience so for instance for our users we want to make it 399 00:31:41,62 --> 00:31:45,15 a flat rate so we pay for everything that you use like as if it is 400 00:31:45,16 --> 00:31:50,42 a paywall but then you pay us once monthly and we can even it out across users just 401 00:31:50,43 --> 00:31:53,78 the same way that like an inside source provider would be like if you use the 402 00:31:53,79 --> 00:31:54,47 Internet more 403 00:31:54,48 --> 00:31:57,25 a service provider has more costs but they're still going to charge you like 404 00:31:57,26 --> 00:32:03,95 a flat rate. I just want to push 405 00:32:03,99 --> 00:32:04,28 back 406 00:32:04,29 --> 00:32:10,05 a little bit on that web site simply just like service ads and you should be able 407 00:32:10,06 --> 00:32:15,14 to like you should have to watch them to your Web site like ad blockers serve 408 00:32:15,15 --> 00:32:19,50 a large lake security purpose you know and usually on 409 00:32:19,51 --> 00:32:22,22 a Web site it's not the actual Web site serving an ad it's like 410 00:32:22,23 --> 00:32:26,57 a third party that is collecting tracking data that sold on some totally 411 00:32:26,58 --> 00:32:33,05 unregulated market. And so like that equals this could just stop all of that but I 412 00:32:33,06 --> 00:32:39,49 want to say it's like kind of creepy to say like you should just watch or add to 413 00:32:39,50 --> 00:32:39,63 that 414 00:32:39,64 --> 00:32:47,61 a lot more than not yet so eager to 415 00:32:47,62 --> 00:32:54,48 be thought about scaling the pay rate for different types of users so for example 416 00:32:54,49 --> 00:33:00,18 if the content is like one hundred seven dollar in this country it might be 417 00:33:00,19 --> 00:33:05,50 affordable for us but not for others in different parts of the country yet so it's 418 00:33:05,51 --> 00:33:08,99 actually kind of interesting because it's it's like that because there's no 419 00:33:09,00 --> 00:33:12,82 operator of this platform it's very much like an interplay between the publishers 420 00:33:12,83 --> 00:33:17,27 and the providers or the users and so like for instance if I'm 421 00:33:17,28 --> 00:33:22,14 a provider I can say I want to I want the browser to pay me and then depending on 422 00:33:22,15 --> 00:33:26,17 how much the browser pays me I will provide this version of the website or that 423 00:33:26,18 --> 00:33:30,23 version of the website right or this experience or that experience and so to employ 424 00:33:30,24 --> 00:33:34,00 in the future you said like that would be up to the publishers to say like hey we 425 00:33:34,01 --> 00:33:37,78 the New York Times believe that content should be accessible to everybody 426 00:33:37,79 --> 00:33:41,35 affordable to everybody so we're going to charge an IP that comes from you know 427 00:33:41,36 --> 00:33:45,37 a poor country less than we charge an IP that comes from the US for example so 428 00:33:45,70 --> 00:33:51,31 publishers could certainly do that OK I think we're out of time thank you. 429 00:34:03,82 --> 00:34:04,06 Does 430 00:34:04,07 --> 00:34:14,07 a great presentation thank you sander. Yeah 431 00:34:14,07 --> 00:34:18,71 . I will give sander and then it to get said that 432 00:34:21,01 --> 00:34:26,17 how are you guys doing today yes dear I think in the Conference Planning talks 433 00:34:26,18 --> 00:34:26,77 naturally. 434 00:34:35,76 --> 00:34:40,31 This is standard and he's going to be talking about textile photos you feel like 435 00:34:40,32 --> 00:34:47,17 you're all set they're all right I want to get away. But I do sort of 436 00:34:47,18 --> 00:34:53,65 a hybrid. Here. So we build textile 437 00:34:53,66 --> 00:34:58,35 photos and we are we say that we're on a mission to build 438 00:34:58,39 --> 00:35:02,41 a really good decentralized user experiences seems like that's kind of 439 00:35:02,42 --> 00:35:04,35 a common seen with 440 00:35:04,36 --> 00:35:09,72 a lot of people the tech cell phone is our first experiment and it is mobile first 441 00:35:09,73 --> 00:35:15,72 offline first photo sharing and back out. If you guys have seen our booth down the 442 00:35:15,73 --> 00:35:21,75 hall we've been running out of it alive shared thread we call them which is 443 00:35:22,46 --> 00:35:23,15 basically just 444 00:35:23,16 --> 00:35:29,78 a shared photo album. To people been coming by and downloading our apps on the App 445 00:35:29,79 --> 00:35:34,95 Store taking photos and they're popping up here peer to peer. 446 00:35:36,78 --> 00:35:43,48 So what is it exactly so essentially for backup and sharing you can post photos 447 00:35:43,54 --> 00:35:48,97 comments chat coming in all the basic stuff you might expect to have like an 448 00:35:48,98 --> 00:35:55,53 Instagram. We want to cover. In Mobile First on the App Store right now we have 449 00:35:55,54 --> 00:36:01,15 a desktop client as well which you see right here so this is our this is showing 450 00:36:01,56 --> 00:36:08,13 a simulator with our i with app and desktop client which is on the decentralized 451 00:36:08,14 --> 00:36:15,13 web thread. Content encrypted and and open source all 452 00:36:15,14 --> 00:36:21,33 of buzz words for you there. And I'm just actually 453 00:36:22,33 --> 00:36:29,24 in. Exit. There is 454 00:36:29,25 --> 00:36:33,41 it can jump to. 455 00:36:38,64 --> 00:36:39,90 Reno to get out of here. 456 00:36:54,67 --> 00:36:55,10 The 457 00:37:03,32 --> 00:37:10,10 shooter is frozen and I think she 458 00:37:10,36 --> 00:37:14,85 is running too many textile nodes. 459 00:37:33,14 --> 00:37:33,28 And. 460 00:37:50,67 --> 00:37:52,43 You get over there. 461 00:38:10,25 --> 00:38:17,04 And you. Well 462 00:38:17,72 --> 00:38:24,46 I guess all this talk through the rest. So. In the last two 463 00:38:24,47 --> 00:38:29,86 sides that are not accessible. Talk about the two main components of 464 00:38:29,87 --> 00:38:36,84 a textile node are our textile project so we run IP FS node from go 465 00:38:36,85 --> 00:38:42,25 IP FS in the wrap it with our application logic. And that's what we call the 466 00:38:42,26 --> 00:38:49,13 textile node and we run. Bridges to 467 00:38:49,17 --> 00:38:51,62 Objective C. And Java bridges which talk to 468 00:38:51,63 --> 00:38:57,20 a reactive library. And we build. 469 00:38:58,27 --> 00:39:04,69 Frameworks with go mobile to talk to them I was an android build. 470 00:39:05,72 --> 00:39:12,72 The main concept of textile is this concept of thread so thread is essentially 471 00:39:12,86 --> 00:39:19,24 the hash chain of hash chain of updates that are committed. They're all indexed by 472 00:39:19,25 --> 00:39:24,91 each node so you can view Cory's and things that any kind of app needs to do. 473 00:39:25,93 --> 00:39:32,17 Each each update on that chain is encrypted with the thread key. And signed by the 474 00:39:32,18 --> 00:39:38,63 peers So for example if I was to send an invite to an existing user on the system. 475 00:39:39,96 --> 00:39:44,04 They would be able to traverse that chain back pick up all the other peers that 476 00:39:44,05 --> 00:39:51,03 have joined the album or thread. And all the content that was came along with 477 00:39:51,04 --> 00:39:57,65 it. And the blocks that are on the chain might contain pointers to 478 00:39:57,69 --> 00:40:00,67 ciphertext and a key to unlock that which is what 479 00:40:00,68 --> 00:40:07,71 a photo is on the system. And 480 00:40:07,72 --> 00:40:12,03 I really wish. I. Had them. 481 00:40:17,05 --> 00:40:23,62 I must have to go to questions I guess anybody. 482 00:40:41,34 --> 00:40:47,20 What's in your missing flight. Well I was mostly just going to show 483 00:40:47,21 --> 00:40:53,03 a demo. Oh 484 00:40:58,17 --> 00:41:02,23 well. I was going to join 485 00:41:02,24 --> 00:41:07,12 a thread and then our other the rest of our team I was going to invite them to the 486 00:41:07,13 --> 00:41:11,58 thread and they're going to start posting photos. So. 487 00:41:15,06 --> 00:41:21,64 Yeah OK. 488 00:42:03,99 --> 00:42:10,18 Yeah. Yeah. Yeah threaded it's 489 00:42:10,19 --> 00:42:15,65 a communication channel so you can put push anything over it and it's persisted. 490 00:42:17,08 --> 00:42:22,76 Conflict free shared state centrally so the first sees case of its photos in our 491 00:42:22,77 --> 00:42:26,81 world we like to do chat and audio and video. 492 00:42:29,92 --> 00:42:36,10 So using IP best for the storage are those files pins or how do you ensure that 493 00:42:36,11 --> 00:42:39,58 they stay up there permanently so you can run 494 00:42:39,59 --> 00:42:45,37 a text on note in Cafe mode that was on the other side. So Cap 495 00:42:45,38 --> 00:42:51,19 a node is very similar to normal text on the it does open user A.P.I. 496 00:42:51,23 --> 00:42:56,49 With with a referral code. And then it will provide 497 00:42:56,53 --> 00:43:02,07 a ping service authenticated with the Web tokens so we want if you download our app 498 00:43:02,08 --> 00:43:07,77 at some app store you will be hitting one of our cafe know that. You could totally 499 00:43:07,78 --> 00:43:11,07 run your own cafe node and you could run it as 500 00:43:11,08 --> 00:43:18,04 a desktop app if you wanted to or put it on Amazon or everyone do. We'd like to 501 00:43:18,08 --> 00:43:24,83 use Falcon at some point Jeremy. What do we need to 502 00:43:25,82 --> 00:43:32,63 be able to use that app now. Download an app store. I'll give you 503 00:43:32,64 --> 00:43:35,66 a referral code suite and Android. 504 00:43:42,98 --> 00:43:49,92 I'm way over so thanks guys. All right son so with alum 505 00:43:50,76 --> 00:43:56,14 coming from Old Dominion University that's in Regina and I'm going to talk about IP 506 00:43:56,15 --> 00:44:03,08 vac everyone was doing something with IP FS I said let's put Wayback Machine in 507 00:44:03,12 --> 00:44:09,95 IP fast and basically replay archive content from IP fs so. 508 00:44:11,72 --> 00:44:14,32 So this introduce what idea first is it's 509 00:44:14,33 --> 00:44:20,41 a location independent file system no matter where the content is at the conduit is 510 00:44:20,42 --> 00:44:27,28 same it's going to you into the very same. Hash and if you had that hash no matter 511 00:44:27,29 --> 00:44:31,86 where the file is if it is in the network you can retrieve that So that's the basic 512 00:44:31,87 --> 00:44:37,95 promise from IP FS Let's move on to how web archives work so there's 513 00:44:37,96 --> 00:44:39,63 a very simple H.T.M.L. 514 00:44:39,64 --> 00:44:46,58 Page this is how the browsers render it to us but this is what programmers write. 515 00:44:50,97 --> 00:44:57,21 But when you. When what browser to get is basically this thing so here is nice 516 00:44:57,22 --> 00:45:03,57 email caught on top of that there is an extra extra to be header the. Robot types 517 00:45:03,58 --> 00:45:10,43 try to archive this whole thing headers and payload so in order to save 518 00:45:10,44 --> 00:45:16,08 that they can have like individual files for that but that will. The choir millions 519 00:45:16,09 --> 00:45:19,67 and billions of files are basically on a file system that will be 520 00:45:19,68 --> 00:45:23,43 a lot to handle and they will be named collisions and so many other problems so 521 00:45:23,44 --> 00:45:29,08 what they do is they take all the terror and payload and extra header on problem 522 00:45:29,09 --> 00:45:34,38 that they call it waters that has more information in it and this whole thing is 523 00:45:34,39 --> 00:45:39,05 called Walk a card and they have many of these records one after the other in 524 00:45:39,06 --> 00:45:44,93 a single file and those files are called files then they index that which you are 525 00:45:44,94 --> 00:45:50,07 all. Archived at what date time is present in which was 526 00:45:50,08 --> 00:45:56,03 a file at what off that and how many bytes from that's how this all basically. So 527 00:45:56,04 --> 00:45:57,05 we changed it 528 00:45:57,06 --> 00:46:03,69 a little bit rather than sitting in walk while the tables are quiet and. Because 529 00:46:03,70 --> 00:46:08,54 that is wanting to have the payload part and head of part and we put it an IP 530 00:46:08,55 --> 00:46:11,56 effort into two different objects and if you asked why two different objects 531 00:46:11,57 --> 00:46:17,16 because header has something different in each response payload can be the same so 532 00:46:17,17 --> 00:46:22,90 we can utilize the. Beautification for that and then it goes into 533 00:46:22,91 --> 00:46:29,02 a professor and we get. Hashes related to those pieces and we put that in our index 534 00:46:29,03 --> 00:46:33,51 basically so we create an index that says this is the you are all going to we are 535 00:46:33,55 --> 00:46:40,00 there at this time and here is the header. Digest and the other one is payload All 536 00:46:40,01 --> 00:46:46,78 right so then in our replay time when someone looks up for that we can locate 537 00:46:46,79 --> 00:46:53,05 the item in our index and identify those hashes fetch that from IP FS 538 00:46:53,33 --> 00:46:59,86 combined them make necessary changes and serve that thing so this is this is how it 539 00:46:59,86 --> 00:47:06,68 works. And so so far we basically only be centralized the storage part of it 540 00:47:06,82 --> 00:47:12,48 not necessarily the index part that we still need that non distributed index in 541 00:47:12,49 --> 00:47:17,67 order to replay that and we have planned for that tomorrow we'll be talking more 542 00:47:17,68 --> 00:47:24,45 details how we can do that so I think that's about it and here is something that I 543 00:47:24,46 --> 00:47:25,34 had a Tribal 544 00:47:25,75 --> 00:47:30,82 a few minutes ago we can go ahead and replay that and this is the size it is 545 00:47:30,86 --> 00:47:35,97 running it was kept about two hours ago. So thank you so much. 546 00:47:41,02 --> 00:47:48,01 I don't. Already I'll take any 547 00:47:48,02 --> 00:47:54,66 questions. Right. 548 00:47:58,93 --> 00:47:59,70 Mine's not actually 549 00:47:59,71 --> 00:48:03,86 a question is yes thank you because we were very inspired by your work in the 550 00:48:04,51 --> 00:48:11,03 project and how old it is how we how rigorous how captured fleeting 551 00:48:11,48 --> 00:48:17,69 We also will say for the whole headers to give Absolutely I mean without that you 552 00:48:17,70 --> 00:48:24,57 cannot recreate. The experience and as of archivists we we we want to 553 00:48:24,58 --> 00:48:27,17 not just a bite but also the experience of it is a lot 554 00:48:27,24 --> 00:48:33,07 a frown on information that you want and so that's why the extra bit in our price 555 00:48:33,08 --> 00:48:37,63 not just the headers that we receive what trying the of the that's not available in 556 00:48:37,64 --> 00:48:39,90 there that's I was going lie about it in that it had 557 00:48:39,91 --> 00:48:43,40 a physical you know. That's 558 00:48:50,07 --> 00:48:52,82 a could you go into how the indexing works 559 00:48:52,83 --> 00:48:59,72 a little short so. Of Currently we do not index and 560 00:48:59,73 --> 00:49:05,76 pushing back as directly instead we basically we take existing was and why we do 561 00:49:05,77 --> 00:49:10,66 live this way because there are so many tools already in the back having community 562 00:49:11,09 --> 00:49:17,03 that create walkways directly so the that there is territory that does that and we 563 00:49:17,07 --> 00:49:21,05 have created our own several tools that are like browser based crawling systems 564 00:49:21,06 --> 00:49:24,72 that basically create walk wild So rather than going and patching all of them and 565 00:49:24,76 --> 00:49:29,25 allowing them to push things to ideas as directly we asked them Hey do what you are 566 00:49:29,26 --> 00:49:34,78 doing create what wild and we will then push content the way we want in IP of 567 00:49:34,79 --> 00:49:39,61 a system and replace their source so once we push in the IP system and create index 568 00:49:39,62 --> 00:49:44,30 out of it we don't need those wildfires but what size are awesome. For long term 569 00:49:44,31 --> 00:49:50,85 provision even though we applied. At the office for replay because Thank you 570 00:49:55,62 --> 00:50:00,82 any more questions we got 571 00:50:00,83 --> 00:50:07,82 a lot of time. All right so the code is available 572 00:50:07,83 --> 00:50:14,72 here to get help and we have sufficiently documented it. How to run it that is some 573 00:50:15,39 --> 00:50:20,55 talk about documentation as well. And actually this is what I'm running right now 574 00:50:21,36 --> 00:50:27,54 the of that your name yes. And this little by noting that you see 575 00:50:27,91 --> 00:50:34,64 Senator think that we added in here and all these. So the vital replay 576 00:50:34,65 --> 00:50:38,32 systems also require something called rewriting So if you have 577 00:50:38,33 --> 00:50:43,83 a web page and it has images and other resources in there they might have like 578 00:50:43,87 --> 00:50:49,23 absolute your eyes or absolute path and what not and based on that your request 579 00:50:49,24 --> 00:50:53,64 might leak from the live event we want to prevent that from happening so if you are 580 00:50:54,21 --> 00:51:00,78 sharing. And I play you want everything in there from our archive nothing from the 581 00:51:00,79 --> 00:51:05,96 lively so to prevent that whereas you really do it like inside what I read about 582 00:51:06,06 --> 00:51:08,02 rewriting so they modify all those U.R.L. 583 00:51:08,03 --> 00:51:12,29 So it doesn't happen but things that were injected using javascript kind of becomes 584 00:51:12,30 --> 00:51:17,21 tricky to to know in advance from the server how to do that and for that we have 585 00:51:17,22 --> 00:51:21,13 some service worker in place here that kind of traps all those requests making 586 00:51:21,14 --> 00:51:25,60 literally modifications before they go it werent kind of bullshit from the bad guy 587 00:51:25,61 --> 00:51:30,19 so that's another interesting piece of code that we have freedom of 588 00:51:33,11 --> 00:51:39,94 all right thank you thank you so much. So I 589 00:51:39,95 --> 00:51:43,57 am going to go kind of higher levels of the super quick talk and then during 590 00:51:43,58 --> 00:51:46,54 questions if you want to I'm kind of pitching them principles that I think are 591 00:51:46,55 --> 00:51:50,46 important for the future of decentralized education and if you think they're 592 00:51:50,47 --> 00:51:54,50 improbable or are interested in knowing more concrete specific 593 00:51:54,54 --> 00:52:00,93 a political ways to make this happen and asking questions so I am a Vancouver D. 594 00:52:00,94 --> 00:52:03,68 Centralist I run with 595 00:52:03,69 --> 00:52:07,93 a lot of weird weird absurdist crowds we just made those combine Coover to the 596 00:52:07,94 --> 00:52:13,61 celebration of radical radical decentralisation and I will be wrapping he night as 597 00:52:13,62 --> 00:52:17,42 news can you actually rapped at the first decentralize Web Summit about and 598 00:52:17,43 --> 00:52:24,00 inexplicably for Richard Stallman which like blew my mind for life so yeah so I was 599 00:52:24,01 --> 00:52:26,02 always a school nerd and it took me 600 00:52:26,03 --> 00:52:29,96 a while to realize the school would not fun for everyone in the way it was for me 601 00:52:29,97 --> 00:52:35,87 and so one of my kind of motivating principles is making education fun for everyone 602 00:52:36,44 --> 00:52:39,30 I've got some pictures here from DOS com We had 603 00:52:39,31 --> 00:52:43,77 a city wide scavenger hunt where we teamed up strangers to go around and do 604 00:52:44,09 --> 00:52:48,64 cryptographic puzzles that is actually they kind of thought this team here thought 605 00:52:48,65 --> 00:52:53,44 that that was some sort of physical puzzle but each of those blocks were set 606 00:52:53,54 --> 00:52:59,49 representing binary code so there are lots of areas of ideation in decentralization 607 00:52:59,60 --> 00:53:04,68 that are not just coding that are kind of game theory ethics etc and we're looking 608 00:53:04,69 --> 00:53:08,51 more and more at design it's coming up all the time so getting different minds in 609 00:53:08,52 --> 00:53:12,42 the room means doing different types of activities to get people excited on the 610 00:53:12,43 --> 00:53:13,22 side there we have 611 00:53:13,23 --> 00:53:18,45 a crypto speed debates between one of our locals met last year and the C.E.O. 612 00:53:18,46 --> 00:53:23,08 Of pied piper coin the participatory thing we kind of threw people into to just 613 00:53:23,09 --> 00:53:24,23 pick a side in 614 00:53:24,24 --> 00:53:30,51 a debate for example we had that connect versus revolt burn but yeah I like the 615 00:53:30,52 --> 00:53:35,36 idea of just letting people run with things they will make will stuff now on the 616 00:53:35,37 --> 00:53:40,45 formal side we actually do need order decentralization. Does not mean disorder and 617 00:53:40,46 --> 00:53:45,24 too often we're like an archaic sovereign citizens as hackers and so there are 618 00:53:45,25 --> 00:53:48,95 these concepts and I'll put these links later on there are these concepts like 619 00:53:48,99 --> 00:53:49,64 a crazy 620 00:53:49,65 --> 00:53:54,49 a self binding setting goals and often the easiest way we understand with this and 621 00:53:54,50 --> 00:53:59,99 we see it come up in crypto economics a lot is skin in the game the link I have on 622 00:54:00,00 --> 00:54:02,25 a credo which I will share out because B.B. 623 00:54:02,26 --> 00:54:06,70 Minder is on point they're like put money up against yourself or habit building 624 00:54:06,71 --> 00:54:09,64 like BET money of very simply and they have 625 00:54:09,65 --> 00:54:14,20 a great write up on just getting yourself to commit not many of us have 626 00:54:14,21 --> 00:54:20,67 a huge amount of intrinsic motivation in the day and age of means in screen and yet 627 00:54:20,68 --> 00:54:25,43 Cory Doctorow did the keno at the first the Web Summit that was on protecting our 628 00:54:25,44 --> 00:54:27,36 future most compromise 629 00:54:27,37 --> 00:54:32,06 a bull selves or are protecting our present idealistic selves from the self that 630 00:54:32,07 --> 00:54:35,04 wants to eat all the cookies at one in the morning so he did 631 00:54:35,05 --> 00:54:37,51 a great write up of his own talk for that but I think we look 632 00:54:37,55 --> 00:54:42,37 a lot of his time at commercial development of platforms and products as like 633 00:54:42,38 --> 00:54:45,90 a scrum Sprint cycle thing and we don't really look at the order we could put into 634 00:54:45,91 --> 00:54:50,78 place for education as well like if I just sign up for if I feel excited about 635 00:54:51,23 --> 00:54:55,31 massive online courses and I sign up for ten of them I'm never going to follow up 636 00:54:55,32 --> 00:54:59,76 but if I set something into place with my peers where we all kind of put some form 637 00:54:59,77 --> 00:55:01,46 of skin in the game together and we have 638 00:55:01,47 --> 00:55:07,87 a syllabus shared that we all commit to then we might follow through there. And I 639 00:55:07,88 --> 00:55:11,62 think the Commons is necessary for that and I have just put means here because I 640 00:55:11,63 --> 00:55:16,85 thing the only real thriving Commons we have right now where people are swapping 641 00:55:16,86 --> 00:55:20,17 and sharing and re mixing is actually in the medical chart and that's why I'm 642 00:55:20,18 --> 00:55:26,85 obsessed with me. And finally I think we don't have rituals in our civilisation as 643 00:55:26,86 --> 00:55:31,73 much anymore specially post religion we don't have coming of age ceremony that 644 00:55:31,74 --> 00:55:35,54 satirise overdosage com We had the State of the decentralized Union in 645 00:55:35,55 --> 00:55:39,49 a UN style building and we all dressed up we're. Didn't like made 646 00:55:39,50 --> 00:55:42,25 a thing of it and it really made this 647 00:55:42,26 --> 00:55:48,15 a the super silly formal time thing like we can play especially in this industry we 648 00:55:48,16 --> 00:55:53,47 can play with the old norms of business and and you know goof around with them 649 00:55:53,48 --> 00:55:56,69 a little bit take some of the formality and play with it because we realize 650 00:55:56,70 --> 00:56:02,32 everyone is pretty playful and here to get me to reset the time just going 651 00:56:04,59 --> 00:56:10,25 so yeah this is something that I think we can all collaboratively design as groups 652 00:56:10,26 --> 00:56:14,82 and I think all the fun silly stuff is just as important as the code because 653 00:56:14,83 --> 00:56:19,48 otherwise we're not going to get normal users excited so yeah if you ever come up 654 00:56:19,49 --> 00:56:25,79 to Vancouver. Come visit us for your Plus an archaic education space we want to 655 00:56:25,80 --> 00:56:30,66 collab with everybody and you can find me online so cool is that the cool guys on 656 00:56:30,67 --> 00:56:33,27 please come in that is my presentation. 657 00:56:38,78 --> 00:56:45,77 Any questions I know it's very abstract so if you don't believe some of these 658 00:56:45,78 --> 00:56:50,85 principles can be made real challenge me we have like we've designed specific 659 00:56:50,86 --> 00:56:53,22 activities. Will I have 660 00:56:53,23 --> 00:57:00,20 a question. How do you do that essentially as well as image is done amazingly 661 00:57:00,21 --> 00:57:03,63 and that's the thing is like the game if occasion and I think I have notes here 662 00:57:03,64 --> 00:57:07,45 that I look at I meant to mention like experience show a game a fight 663 00:57:07,46 --> 00:57:11,06 a learning and stuff from the ferry first day you know we did like the courtyard 664 00:57:11,07 --> 00:57:15,17 activities and there's like the awesome art puzzle themself like that I think of 665 00:57:15,18 --> 00:57:20,53 keeping everyone refresh giving people some different ways to still be learning but 666 00:57:20,54 --> 00:57:24,87 not just kind of like whoa information jam information damn thing off I'm. 667 00:57:30,42 --> 00:57:34,26 On to another one. Yes it's 668 00:57:34,27 --> 00:57:40,43 a do is clean millionaire. How can we how can we take the fun that we've created 669 00:57:40,44 --> 00:57:47,02 offline back online because it's also really easy to fall into serious pants mood 670 00:57:47,03 --> 00:57:53,77 online too that's very true and that is why I very much will lean on me and gifts I 671 00:57:53,78 --> 00:58:00,09 mean enlightening talks yesterday we got into the area of see if. You can pronounce 672 00:58:00,10 --> 00:58:05,86 it anyway but I feel honestly that finding ways for humor and art to break up the 673 00:58:05,87 --> 00:58:10,04 monotony and maybe even if we started to establish cultural norms around like oh 674 00:58:10,05 --> 00:58:13,66 this debate on crypto Twitter is getting way to eat it I think everybody needs to 675 00:58:13,67 --> 00:58:17,51 drop their most fun. For 676 00:58:17,52 --> 00:58:22,73 a second it kind of like humor humor coming and I think always will pull us back 677 00:58:22,74 --> 00:58:29,58 from excessive drama. Can you touch base on. A 678 00:58:29,59 --> 00:58:31,14 topic or expand on 679 00:58:31,15 --> 00:58:35,75 a topic that I've heard you speak about before about how the decentralization that 680 00:58:35,76 --> 00:58:41,81 decontrol helps too. Prevent the 681 00:58:41,82 --> 00:58:47,80 commoditization nuke decontrol Oh hey out OK so our space in Vancouver the reason 682 00:58:47,81 --> 00:58:48,12 we are 683 00:58:48,13 --> 00:58:55,06 a four plus year old community common space and you haven't heard of us is because 684 00:58:55,07 --> 00:59:00,38 we're this weird gross little basement with no real official appeal affiliations 685 00:59:00,39 --> 00:59:04,73 and it's been beautiful because business people come down during the you know the 686 00:59:04,74 --> 00:59:06,58 exciting either twenty K. 687 00:59:06,59 --> 00:59:10,71 Burst in December and stuff like that and they're like hey do I don't want to sit 688 00:59:10,72 --> 00:59:15,02 down here like it was that sticky and intentionally protected that like us being 689 00:59:15,03 --> 00:59:16,20 a bunch of goofballs renting 690 00:59:16,21 --> 00:59:20,71 a basement together has really protected us from getting exploited so I almost feel 691 00:59:20,72 --> 00:59:22,30 like levity is also 692 00:59:22,31 --> 00:59:27,59 a way like this levity of the goofing around dressing silly etc It's almost 693 00:59:27,60 --> 00:59:34,44 a way to sneak past like the sociopathic singular focus of like Wall Street style 694 00:59:34,45 --> 00:59:38,64 mentality is they're like oh I can't pay attention to these goofs they're all in 695 00:59:38,65 --> 00:59:42,26 costumes they were like it's kind of like in walking dead when they like all wrap 696 00:59:42,27 --> 00:59:47,47 themselves in zombies and they go through it I guess kind of the exact opposite but 697 00:59:47,71 --> 00:59:51,13 it is something that I really think like I think we've been playing a game for 698 00:59:51,14 --> 00:59:55,72 a long time where like I don't think anyone wants to be that serious and people 699 00:59:55,73 --> 01:00:00,22 kind of were doing this cultural game where we all thought we had to wear suits and 700 01:00:00,23 --> 01:00:04,45 that's changing significantly so I think we can keep keep with that trend. 701 01:00:09,69 --> 01:00:16,65 I want to talk about collaborative creation of educational material. The 702 01:00:16,69 --> 01:00:23,42 courseware companies cities so I don't accept updates. If you 703 01:00:23,43 --> 01:00:28,56 have some ways of sharing a translation but it would seem like there is a kind of 704 01:00:28,57 --> 01:00:33,13 a quality issue that you would you would like to sever courseware that is produced 705 01:00:33,53 --> 01:00:34,73 and there's a good quality is 706 01:00:34,74 --> 01:00:41,27 a good documentary but those those are the average enterprises that are put 707 01:00:41,28 --> 01:00:47,45 together for millions of dollars to do other good quality documentary 708 01:00:48,14 --> 01:00:51,84 and yet the course very still suffers from being 709 01:00:51,85 --> 01:00:58,29 a kind of replication of the art of the in classroom experience Sim City and 710 01:00:58,99 --> 01:01:04,62 yeah I mean this in this many think I think we need to follow the path 711 01:01:05,49 --> 01:01:09,91 blazed by open source software communities if you have like some of the best 712 01:01:09,92 --> 01:01:14,82 curated resource lists for learning about watching like Jamieson lobs guide and 713 01:01:14,83 --> 01:01:20,20 stuff like that are run off of people submitting Paul requests so if we have 714 01:01:20,39 --> 01:01:24,01 platforms where our peers can comment and say hey I don't think that's 715 01:01:24,02 --> 01:01:27,70 a great resource and we have an open discussion always going on around it and it 716 01:01:27,71 --> 01:01:31,17 can be revised commented upon I think that's 717 01:01:31,18 --> 01:01:38,14 a good way for us to keep upping our level of shared knowledge Excellent thank you 718 01:01:38,15 --> 01:01:39,06 so much so C. 719 01:01:39,07 --> 01:01:45,88 For that super entertaining time thank you. That concludes the final 720 01:01:45,89 --> 01:01:49,98 lightning talk session that we have today I want to give 721 01:01:49,99 --> 01:01:56,98 a huge thanks to Amy for hosting this one and to all of the rest of the 722 01:01:57,02 --> 01:02:03,14 hosts who volunteered to help thank you too to the volunteers who have been around 723 01:02:03,15 --> 01:02:08,57 the room helping to keep keep everything running on schedule. 724 01:02:10,17 --> 01:02:15,99 You sayings to all of the speakers here in the previous sessions thanks to the you 725 01:02:16,00 --> 01:02:22,80 to live audience for tuning in online. In the back like huge 726 01:02:22,81 --> 01:02:29,65 shout out to you guys for being awesome. The staff here at the Mint the 727 01:02:29,66 --> 01:02:33,96 free coffee the security guards they're all doing a great job this was 728 01:02:33,97 --> 01:02:38,97 a really amazing event and so finally you know huge thanks to the Internet Archive 729 01:02:39,01 --> 01:02:44,70 for helping to put it on the sponsors and of course to you the lovely audience for 730 01:02:44,71 --> 01:02:51,66 coming in giving someone to talk to people who care about decentralization 731 01:02:51,82 --> 01:02:56,69 and really understanding that decentralization is a means to an end which is 732 01:02:57,39 --> 01:03:02,78 a better web for all of us thank you for coming and participating in the 733 01:03:02,79 --> 01:03:03,91 centralized Web Summit.